Multi-objective Decision in Machine Learning

被引：6
作者：
de Medeiros T.H. [1 ]
Rocha H.P. [2 ,5 ]
Torres F.S. [3 ,5 ]
Takahashi R.H.C. [4 ,5 ]
Braga A.P. [3 ,5 ]
机构：
[1] Department of Computing and Systems, Federal University of Ouro Preto, Joao Monlevade, MG
[2] Institute of Engineering Science and Technology, Federal University of Jequitinhonha and Mucuri Valleys, Janaúba, MG
[3] Department of Electronics Engineering, Federal University of Minas Gerais, Belo Horizonte, MG
[4] Department of Mathematics, Federal University of Minas Gerais, Belo Horizonte, MG
[5] Graduate Program in Electrical Engineering, Federal University of Minas Gerais, Av. Antnio Carlos 6627, Belo Horizonte, 31270-901, MG
来源：
Braga, Antônio Pádua (apbraga@ufmg.br) | 1600年 / Springer Science and Business Media, LLC卷 / 28期
关键词：
Classification; Decision-making; Machine learning; Multi-objective optimization;
D O I：
10.1007/s40313-016-0295-6
中图分类号：
学科分类号：
摘要：
This work presents a novel approach for decision-making for multi-objective binary classification problems. The purpose of the decision process is to select within a set of Pareto-optimal solutions, one model that minimizes the structural risk (generalization error). This new approach utilizes a kind of prior knowledge that, if available, allows the selection of a model that better represents the problem in question. Prior knowledge about the imprecisions of the collected data enables the identification of the region of equivalent solutions within the set of Pareto-optimal solutions. Results for binary classification problems with sets of synthetic and real data indicate equal or better performance in terms of decision efficiency compared to similar approaches. © 2016, Brazilian Society for Automatics--SBA.
引用
收藏
页码：217 / 227
页数：10
相关论文
共 26 条
  • [1] UCI machine learning repository., (2007)
  • [2] Barroso M., Takahashi R.H.C., Aguirre L., Multi-objective parameter estimation via minimal correlation criterion, Journal of Process Control, 17, 4, pp. 321-332, (2007)
  • [3] Bland R.G., Goldfarb D., Todd M.J., The ellipsoid method: A survey, (1980)
  • [4] Chankong V., Haimes Y., Optimization-based methods for multiobjective decision-making: An overview, Large Scale Systems, 5, pp. 1-33, (1983)
  • [5] Collette Y., Siarry P., Multiobjective optimization: Principles and case studies. Decision engineering, (2003)
  • [6] Cortes C., Vapnik V., Support-vector networks, Machine Learning, 20, pp. 273-297, (1995)
  • [7] Costa M.A., Braga A.P., de Menezes B.R., Improving generalization of mlps with sliding mode control and the Levenberg–Marquardt algorithm, Neurocomputing, 70, 79, pp. 1342-1347, (2007)
  • [8] Das I., Dennis J.E., A closer look at drawbacks of minimizing weighted sums of objectives for pareto set generation in multicriteria optimization problems, Structural and Multidisciplinary Optimization, 14, pp. 63-69, (1997)
  • [9] Evgeniou T., Pontil M., Poggio T., Regularization networks and support vector machines, Advances in Computational Mathematics, 13, 1, pp. 1-50, (2000)
  • [10] Geman S., Bienenstock E., Doursat R., Neural networks and the bias/variance dilemma, Neural Computation, 4, 1, pp. 1-58, (1992)
123